AI-speed attacks: The line that matters most here: criminals used AI agents to break into 400+ banks. Attackers now move at machine speed. If your code only gets reviewed once a year at audit time, you’re defending against 2026 with a 2015 process.
The internet is going into a LOCKDOWN.
Hundreds of millions of websites are potentially at risk from cyberattacks.
Governments, banks and major platforms have already taken their websites OFFLINE.
Many are already calling it “Black October.”
At the centre of the shutdowns is something most of us have never heard of, but use every day.
PHP, the software behind roughly 7 in 10 websites, has two critical flaws that can let hackers break in without a password.
Dubbed PHP2Shell, both are rated 9.2 out of 10 for severity.
Details began circulating on Oct 20. Emergency fixes arrived the next day, but attackers were already moving.
Cloudflare’s CEO says hacking attempts then jumped from thousands to MILLIONS within hours.
Cybersecurity agencies across several countries are telling PHP site owners the same thing: Patch it now, or take it offline.
Websites have gone dark worldwide, from government portals and hospitals to banks.
CrowdStrike warns these attacks leave almost no trace, so some sites may not know if they’ve been hit.
Around 2.6 MILLION internet-facing servers run PHP, potentially hosting hundreds of sites each.
Over 1.1 million use older versions no longer receiving official PHP updates.
Older installations may need an upgrade to a supported PHP version or a fix from their software provider.
And patching does not remove a backdoor an attacker may already have planted.
Cloudflare also says hackers are now tweaking their attacks to try to slip past its protections.
But PHP isn’t the only security story unfolding.
-Japan’s Onamae suspended ALL Desktop Cloud plans after malware was found on customer servers.
-India’s eMudhra had 8 website-certificate keys so weak, one was literally the number “4”.
-State Bank of India confirmed 3,897 customer records on the dark web.
-Criminals used AI agents to break into 400+ Latin American banks.
-Capcom confirmed a breach exposing 11.7 million email addresses.
-GitLab disclosed a 10/10 severity flaw in versions going back to 2014.
-LiteLLM removed a malware-containing release. Railway reported possible exposure of customer secrets.
None of these has a confirmed link to PHP2Shell.
We’re seeing separate breaches unfold together.
The true number of incidents could take months to come out.
The world is still on alert. Attacks are still coming.
Patch PHP to 8.5.12, 8.4.27, 8.3.36 or 8.2.35 today, or take it offline like NSA says. Then ask the harder question: when did someone last review the code you wrote on top of it?
Effective Immediately, all organizations using PHP for their systems are advised to be prepared for an update addressing two critical vulnerabilities. PHP is widely used for web applications. These vulnerabilities often provide a direct path to the server's file system or database. If an immediate update is not feasible, it is recommend that organizations take the system offline until it can be fully mitigated. NSA will follow up shortly with more specific guidance.
"We got audited" might be the most dangerous sentence in crypto.
It means your code was safe on one specific day.
What have you merged since?
43 vulnerabilities. 2 High Severity. Zero manual review.
The Tribunal: Hunter, Skeptic, Judge.
It caught what a snapshot audit would've missed until it was too late.
This is what continuous security looks like when it's not a checkbox.
safestackai.com
43 findings. 2 High Severity. One scan.
That's what SafeStackAI found on a protocol that had already passed a manual audit.
The gap between "audited" and "secure" is bigger than most teams think.
Most AppSec stacks are a pile of scanners shouting at each other. 900 findings, 12 real.
Posture doesn't improve by adding another alert. It improves when something actually reasons about which ones matter and proves it.
market_state: bullish
shipping_velocity: up
audit_frequency: unchanged
that gap is where exploits live.
A vulnerability sat in Balancer V2 for 4 years.
Introduced 10 days after their final audit. Exploited November 2025 for $120M.
We ran it through SafeStackAI. All 5 critical components flagged.
Audits are a snapshot. Code doesn’t stop shipping after they end.
Ran our pipeline on a protocol that had already been audited.
Found 43 issues. 2 HIGH.
Audits are point in time. Code isn't.
Just shipped Move support. 26 Sui-native vulnerability classes, full parity with our Solidity, Go, Rust, JS/TS, and Python coverage. One pipeline, six languages, zero excuses.
SafeStackAI retweeted

Proofable@proofablemeJul 17
🇺🇸
United StatesConnected via United States App StoreAccount-level information, not a live location or per-post device.
This is the gap we built SafeStackAI to close. An inherited initializer left callable, a deploy script that masks it in the same tx, nothing here trips a regex scanner. Exactly the kind of finding continuous, reasoning-based coverage is for.
.@SafeStackAI found YieldcoinShare had 2 initialize(). @chainlink ACE ComplianceTokenERC3643 initialize was still present and not overridden.
Deploy script calls initialize in same tx so doubtful it'd be frontrun. Maybe that's why it wasnt flagged by other tools (?)
SafeStackAI retweeted
The wait is over. Today is the day. 🚀
After months of offline focus, StudIQ is back, and we're locking down campus security.
Join us and @nickbronsdoteth from @SafeStackAI to hack-proof your hustle.
📅 Today, 6:00 PM WAT 📍 Register: studiq.fun/Webinar
#StudIQ
The exams are done, the silence is over, and we are officially BACK. 🚀
We’re kicking off our return with @nickbronsdoteth (Founder, @SafeStackAI ) to show you how to secure your capital and hack-proof your online hustle.
📅 Fri, July 17 | 6PM WAT
📍 Google Meet
Building on @SuiNetwork? We'd love to put the Tribunal on your codebase.
Free tier available, takes minutes to connect: safestackai.com
@SuiNetwork ships fast, and fast shipping means fast-growing audit gaps.
Continuous coverage closes them: every commit gets the audit treatment, with findings mapped to STRIDE and OWASP.
Why this matters: audits made Web3 feel safe once. But every PR merged after audit day is unreviewed attack surface.
@Balancer's $120M exploit was introduced 10 days after their final audit. It sat in production for 4 years.
Every finding runs through the Tribunal, our adversarial three-agent pipeline.
The Hunter finds it.
The Skeptic tries to disprove it.
The Judge issues a verdict.
Not a hedge, a verdict.
Zero hallucinated findings by design.
This is not a linter port.
We built a Move-specific vulnerability knowledge base from scratch: 26 @SuiNetwork vulnerability classes across 7 categories, each with instructions for how to find it, test it, and validate it.











